Ars Technica NewsDec 2
Syntax hacking: Researchers discover sentence structure can bypass AI safety rules https://arstechni.ca/nRMq #NortheasternUniversity #spuriouscorrelations #largelanguagemodels #VinithM.Suriyakumar #promptinjections #machinelearning #ChantalShaib #jailbreaking #AIalignment #AIresearch #AIsecurity #AIstudy #Biz&IT #GPT-4o #openai #meta #OLMo #MIT #AI
Syntax hacking: Researchers discover sentence structure can bypass AI safety rules

New research offers clues about why some prompt injection attacks may succeed.

Ars Technica